[gentoo-dev] slang v.s. ncurses

[gentoo-dev] slang v.s. ncurses

[Georgi Georgiev]

This is not supposed to start a flame-fest, but I need some advice.

As of mutt-1.5.10-r1, the slang use flag is ignored and ncurses is used
instead. After checking out bugs #96603 #102558 #57416 (mentioned in the
ChangeLog as the reasoning behind the no-slang decision) I got some idea
as to why -- ncurses provides more features, slang was only supposed to
be a minimalist replacement for ncurses, etc.

However, I am used to the control-s shortcut to send spam to a folder,
which spamassassin learns from.

^S          macro                  s=spam-learn\r

This magic key sequence works fine with my slang-linked mutt, but it
does not with a ncurses-linked mutt. I am aware what Control-S is
supposed to do historically. I did some googling around and found out
that people were having similar problems with emacs where ^s is used for
searching. Unfortunately, all the advice they got was "use a different
shortcut mapping". Well, I *will* do that, eventually, but is there no
way to get Control-S working with ncurses? I don't care about the
historical reasons for ^s, and I did have it working until recently. I
feel cheated by the switch to ncurses in this respect.

-- 
(*   Georgi Georgiev   (* Q: What do Winnie the Pooh and John the      (*
*)    chutz@gg3.net    *) Baptist have in common? A: The same middle   *)
(*  +81(90)2877-8845   (* name.                                        (*
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] slang v.s. ncurses

[Tavis Ormandy]

--On Monday, September 05, 2005 14:41:45 +0900 Georgi Georgiev
<chutz@gg3.net> wrote:
> key sequence works fine with my slang-linked mutt, but it
> does not with a ncurses-linked mutt. I am aware what Control-S is
> supposed to do historically.

stty stop undef

-- 
-------------------------------------
taviso@sdf.lonestar.org | finger me for my gpg key.
-------------------------------------------------------

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] slang v.s. ncurses

[Georgi Georgiev]

maillog: 05/09/2005-07:47:17(+0100): Tavis Ormandy types
> 
> 
> --On Monday, September 05, 2005 14:41:45 +0900 Georgi Georgiev
> <chutz@gg3.net> wrote:
> > key sequence works fine with my slang-linked mutt, but it
> > does not with a ncurses-linked mutt. I am aware what Control-S is
> > supposed to do historically.
> 
> stty stop undef

Fair enough. Thanks for the tip.

-- 
 /   Georgi Georgiev    / Bacon's not the only thing that's cured by    /
\     chutz@gg3.net    \  hanging from a string.                       \
 /  +81(90)2877-8845    /                                               /
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] slang v.s. ncurses

[Thomas Dickey]

[-- Attachment #1: Type: text/plain, Size: 614 bytes --]

On Mon, Sep 05, 2005 at 07:47:17AM +0100, Tavis Ormandy wrote:
> 
> 
> --On Monday, September 05, 2005 14:41:45 +0900 Georgi Georgiev
> <chutz@gg3.net> wrote:
> > key sequence works fine with my slang-linked mutt, but it
> > does not with a ncurses-linked mutt. I am aware what Control-S is
> > supposed to do historically.
> 
> stty stop undef

that's simple.  I would have pointed out that ncurses can use a different
mode (start by "man raw"), but that mutt is using a particular mode presumably
for a good reason.

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net

[-- Attachment #2: Type: application/pgp-signature, Size: 228 bytes --]

Re: [gentoo-dev] slang v.s. ncurses

[Georgi Georgiev]

[-- Attachment #1: Type: text/plain, Size: 961 bytes --]

maillog: 05/09/2005-06:16:16(-0400): Thomas Dickey types
> On Mon, Sep 05, 2005 at 07:47:17AM +0100, Tavis Ormandy wrote:
> > 
> > 
> > --On Monday, September 05, 2005 14:41:45 +0900 Georgi Georgiev
> > <chutz@gg3.net> wrote:
> > > key sequence works fine with my slang-linked mutt, but it
> > > does not with a ncurses-linked mutt. I am aware what Control-S is
> > > supposed to do historically.
> > 
> > stty stop undef
> 
> that's simple.  I would have pointed out that ncurses can use a different
> mode (start by "man raw"), but that mutt is using a particular mode presumably
> for a good reason.

Yes, the cbreak() mode allows for ^C and ^Z to work, which doesn't
happen with raw().

I'm glad that there is at least a workaround.

-- 
/    Georgi Georgiev   /  Don't mind him; politicians always sound     /
\     chutz@gg3.net    \  like that.                                   \
/   +81(90)2877-8845   /                                               /

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]